https://sourceware.org/bugzilla/show_bug.cgi?id=32491
--- Comment #9 from Thiago Jung Bauermann <thiago.bauermann at linaro dot org>
---
(In reply to Thiago Jung Bauermann from comment #6)
> Maybe the gas change surfaced a latent bug in ld?
It actually surfaced a latent bug in GCC's configure script: the fact that
commit d5cbf916be4a ("gas/ELF: also reject merge entity size being zero")
caused gas to be stricter about mergeable sections without an entity size
meant that GCC's configure had to use a fallback to detect whether the
assembler supports section merging, and the fallback didn't work.
This made GCC avoid putting debug strings in the .debug_str section, which
caused it to generte a begin.o file with an empty .debug_str section. This
in turn prompted ld to complain about an "invalid operation" when asked to
compress an empty section.
IOW an improvement in gas caused GCC to change behaviour, leading to an
error in ld. Rube Goldberg would be proud. 🙂
I just posted a fix here:
https://inbox.sourceware.org/gcc-patches/20241227214756.1059146-1-thiago.bauerm...@linaro.org/
--
You are receiving this mail because:
You are on the CC list for the bug.
